index
:
ampere-computing/llvm.git
release_60-f1b37feef3d-amp-20180630
release_70-e8af9b4c407-amp-20181130
LLVM including Ampere Computing toolchain specific patches
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
test
/
Transforms
/
LoopLoadElim
Age
Commit message (
Expand
)
Author
2017-01-27
[PM] Port LoopLoadElimination to the new pass manager and wire it into
Chandler Carruth
2017-01-06
Fix LoopLoadElimination to keep original alignment on the inital hoisted store
Mehdi Amini
2016-06-28
[LLE] Don't hoist conditionally executed loads
Adam Nemet
2016-06-17
[LAA] Enable symbolic stride speculation for all LAA clients
Adam Nemet
2016-06-17
[LLE] Don't hard-code the name of the preheader in test
Adam Nemet
2016-06-16
[LLE] New test to check that no versioning for symbolic strides occurs. NFC
Adam Nemet
2016-03-24
[LLE] Check for mismatching types between the store and the load earlier
Adam Nemet
2016-03-10
[LLE] Add missed LoopSimplify dependence
Adam Nemet
2016-03-09
[LLE] Add missing check for unit stride
Adam Nemet
2016-03-01
[LLE] Add testcase for the fix in r262267
Adam Nemet
2016-02-05
[LoopLoadElim] Don't allow versioning when optForSize
Adam Nemet
2016-02-05
Fix typo in comment
Adam Nemet
2015-11-09
Allow LLE/LD and the loop versioning infrastructure to use SCEV predicates
Silviu Baranga
2015-11-03
LLE 6/6: Add LoopLoadElimination pass
Adam Nemet